Re: This is cool.

On Tue, Aug 1, 2017 at 10:36 AM, Mark Murphy/STAR BASE Consulting Inc.
<mmurphy@xxxxxxxxxxxxxxx> wrote:

I don't know how this is wallowing in the past.

Not that this was your point, but I would like to remind people that
newer isn't *always* better. We should be thinking critically, instead
of using chronological age as a proxy for quality (or lack thereof).

It all seems a bit arbitrary to me. Particularly since [various methods of initialization] all seem equally valid in my mind.

Indeed. In addition to what you mentioned, it's also perfectly
reasonable to want your *type declarations* separate from any
*assignment of data*. Even if that assignment is at initialization
time. And, all else being equal, I can totally see how it might be
nice to have all your initializations in one place, rather than some
at the top (in D specs) and some at the bottom.

(I'm not advocating that we now do ALL our initializations in *INZSR.
Just adding to Mark's theme of "there are multiple reasonable ways" to
do things.)

John Y.